Regional variation in stand structure and development in ...

Regional variation in stand structure and development in forests of Oregon, Washington, and inland Northern California

Reilly, M. J., & Spies, T. A. (2015). Regional variation in stand structure and development in forests of Oregon, Washington, and inland Northern California. Ecosphere, 6(10), 192. doi:10.1890/ES14-00469.1

10.1890/ES14-00469.1

Ecological Society of America

Version of Record

http://cdss.library.oregonstate.edu/sa-termsofuse

http://survey.az1.qualtrics.com/SE/?SID=SV_8Io4d9aAYR1VgGx

http://cdss.library.oregonstate.edu/sa-termsofuse

Regional variation in stand structure and developmentin forests of Oregon, Washington, and inland Northern California

MATTHEW J. REILLY1,� AND THOMAS A. SPIES2

1Department of Forest Ecosystems and Society, College of Forestry, Oregon State University, Corvallis, Oregon 97331 USA2United States Department of Agriculture, Forest Service, Pacific Northwest Research Station, Corvallis, Oregon 97331 USA

Citation: Reilly, M. J., and T. A. Spies. 2015. Regional variation in stand structure and development in forests of Oregon,

Washington, and inland Northern California. Ecosphere 6(10):192. http://dx.doi.org/10.1890/ES14-00469.1

Abstract. Despite its importance to biodiversity and ecosystem function, patterns and drivers of

regional scale variation in forest structure and development are poorly understood. We characterize

structural variation, create a hierarchical classification of forest structure, and develop an empirically based

framework for conceptualizing structural development from 11,091 plots across 25 million ha of all

ownerships in Oregon, Washington, and inland Northern California, USA. A single component related to

live tree biomass accounted for almost half of the variation in a principal components analysis of structural

attributes, but components related to live tree density and size, dead wood, and understory vegetation

together accounted for as much additional variation. These results indicate that structural development is

more complex than a monotonic accumulation of live biomass as other components may act independently

or emerge at multiple points during development. The classification revealed the diversity of structural

conditions expressed at all levels of live biomass depending on the timing and relative importance of a

variety of ecological processes (e.g., mortality) in different vegetation zones. Low live biomass structural

types (,25 Mg/ha) illustrated the diversity of early-seral conditions and differed primarily in density of live

trees and abundance of snags and dead wood. Moderate live biomass structural types (25–99 Mg/ha)

differed in tree size and density and generally lacked dead wood, but some structurally diverse types

associated with partial stand-replacing disturbance had abundant live and dead legacies. High live

biomass structural types (.100 Mg/ha) substantiated the diversity of later developmental stages and

exhibited considerable variation in the abundance of dead wood and density of big trees. Most structural

types corresponded with previously described stages of development, but others associated with

protracted early development, woodland/savannah transitions, and partial stand-replacing disturbance

lacked analogs and indicated alternative pathways of development. We propose a conceptual framework

that distinguishes among families of pathways depending on the range of variation along different

components of structure, the relative importance of different disturbances, and complexity of pathways.

Our framework is a starting point for developing more comprehensive models of structural development

that apply to a wider variety of vegetation zones differing in environment and disturbance regimes.

Key words: disturbance history; early-seral vegetation; forest structural development; live and dead biomass; logging;

old-growth; Pacific Northwest; tree density and size; understory vegetation; wildfire.

Received 25 November 2014; revised 6 March 2015; accepted 18 March 2015; published 28 October 2015. Corresponding

Editor: D. P. C. Peters.

Copyright: � 2015 Reilly and Spies. This is an open-access article distributed under the terms of the Creative Commons

Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the

original author and source are credited. http://creativecommons.org/licenses/by/3.0/

� E-mail: [email protected]

v www.esajournals.org 1 October 2015 v Volume 6(10) v Article 192

INTRODUCTION

Despite the importance of forest structure toecosystem function (Waring and Running 2007)and biodiversity (MacArthur and MacArthur1961), variation in structure at the regional scaleis poorly understood. Past regional scale studieshave focused primarily on how climate con-strains biomass (Gholz 1982, Malhi et al. 2006,Urquiza-Haas et al. 2007, Hudiburg et al. 2009),but more detailed investigations of other struc-tural attributes such as standing dead trees(snags) and dead or downed wood are rare(Spies and Franklin 1991, Ohmann and Waddell2002, Ohmann et al. 2007, Ares et al. 2012).Although structural development has been con-ceptualized for a few well-studied forest typesfollowing stand-replacing disturbance (e.g., Bor-mann and Likens 1979, Oliver and Larson 1990,Franklin et al. 2002), we lack a conceptually orempirically based biogeographic understandingof how variation in the distribution of live anddead biomass results in different structuralconditions and developmental pathways acrossregional (.100,000 ha) gradients in climate,disturbance regimes, and species composition.

Vegetation structure can be broadly defined asthe vertical and horizontal distribution andarrangement of live and dead vegetation (Spies1998). In most temperate forest ecosystems,structure can be broken down into three majorcomponents including live trees, dead wood, andnon-arborescent understory plants (Franklin etal. 2002). The live tree component is the mostcommonly studied component of forest structureand is often described using simple metrics (e.g.,biomass, basal area, tree density) as well asvariation in size and spatial arrangement of trees(e.g., Pommerening 2002). The dead woodcomponent can be broken down into the numberand size of snags and dead and downed wood.Snag abundance is often quantified by density(per unit area), but because dead wood iscontinually breaking down through fragmenta-tion and decomposition, biomass estimates offermore precise measures of the abundance of snagsand dead and downed wood in ecosystemstudies (Harmon et al. 1986). Understory vege-tation can be further characterized into functionalgroups (e.g., forbs, shrubs, graminoids) and isgenerally quantified with estimates of cover or

biomass. All three components are intimatelylinked as the cumulative result of the processes ofgrowth, mortality, and decomposition (Spies1998). As a result, forest structure is an inherentlycomplex multivariate concept with broad ecolog-ical implications that vary by individual compo-nents and ecological context.

Much of our current conceptualization ofstructural variation at the stand scale (1–100 ha)has focused on structural development followingstand-replacing disturbances such as wildfireand logging (e.g., Bormann and Likens 1979,Oliver and Larson 1990, Franklin et al. 2002).These conceptual models are founded on a linearage- and/or process-based framework of standdevelopment. Chronological stages proceedalong a single pathway and represent uniquestructural conditions resulting from the domi-nant endogenous processes occurring duringeach stage (e.g., stand initiation/ reorganization,stem exclusion/self-thinning, understory re-initi-ation/ maturation, old-growth). Forest structuremay be simple or complex during early devel-opmental stages depending on disturbance typeand the presence of biological legacies such aslarge live trees, snags, and downed wood(Franklin et al. 2002, Swanson et al. 2010, Donatoet al. 2012). Structure during mid developmentalstages is generally considered less diverse sincelive trees are typically dominated by a singlecohort and the dead wood and snags created bystand-replacement disturbances have largelydecomposed (Spies et al. 1988). Structural diver-sity increases during later developmental stageswhen a variety of live and dead tree sizes, as wellas dead wood on the forest floor, are all present(Spies 1998).

Although these models are useful for concep-tualizing structural development in forests sub-ject to stand-replacing disturbances, they aretypically idealized, based on theory, and haveyet to be evaluated against large empirical datasets of existing forest conditions across a regionalextent including a variety of vegetation zones.Multiple, non-linear developmental pathwaysare possible in vegetation zones where distur-bances operate at a range of severities (Frelich2002). Frequent low-severity disturbances havethe potential to maintain structure (Platt et al.1988) while moderate-severity, partial stand-replacing disturbances may accelerate structural


REILLY AND SPIES

development (Veblen et al. 1991). Additionally,some forests may develop from non-forestedstates including grasslands and shrublands (e.g.,Archer 2010) and savannahs and woodlands maybe intermediate stages between states. Althoughit is evident that a variety of structural conditionsin early (Halpern 1988, Donato et al. 2012) andlater developmental stages are possible within asingle vegetation zone (McCune and Allen 1985),we lack a generalizable framework to describethe diversity of potential pathways of structuraldevelopment across multiple vegetation zones.

Forest structure is a major component ofbiological diversity (Spies 1998, McComb 2008)and focus of regionally based planning efforts(USDA Forest Service, DOI, and BLM 1994).Maintaining, creating, and restoring particularelements of forest structure have been much ofthe focus of coarse and meso filter strategies(Hunter 2005) pertaining to the conservation ofbiological diversity in many parts of the world(Australian and New Zealand EnvironmentalConservation Council 2000, U.S. Fish and Wild-life Service 2003, 2011, Montreal Process WorkingGroup 2009). Despite the importance of foreststructure as a major component of habitatdiversity (McComb et al. 1993, Franklin andVan Pelt 2004, Verschuyl et al. 2008) andapplication of forest structure types in foreststate and transition models and wildlife man-agement guides (Johnson and O’Neil 2001,Hemstrom et al. 2004), we are not aware of anyempirically based regional scale characterizationsof stand structure. Developing an understandingof regional scale variation is important forconservation planning and monitoring, as wellas for developing advanced theoretical modelsand frameworks capable of quantifying regionalforest dynamics in response to a warming climateand altered disturbance regimes.

We used a regional forest inventory represent-ing over 11,000 plots to characterize foreststructure on lands of all ownerships across 25million ha of forested land in Oregon, Wash-ington, and inland Northern California to ad-dress the following questions:

1. What are the major components of foreststructure at a regional scale?

2. What unique structural conditions arecreated from different combinations of

individual structural components?3. How do these unique structural conditions

differ in age in different vegetation zones?4. How do empirically defined developmental

stages relate to those in theoretical models ofstructural development?

5. How do potential pathways of structuraldevelopment differ among vegetationzones?

METHODS

Study regionOur study region is approximately 25 million

ha and includes all forest lands in Oregon,Washington, and inland Northern California(Fig. 1). The region is highly diverse and includesa variety of vegetation zones that follow broadclimatic and topographic gradients (Franklin andDyrness 1973, Barbour and Major 1988, Ohmannand Spies 1998). The climate is generally medi-terranean with most precipitation falling in thewinter (though some portions of the eastern partof the region receive a large proportion insummer thunderstorms), but large gradients inprecipitation, temperature, and elevation create awide range of climatic conditions from warm andmoist at low elevations near the coast to cold anddry at higher elevation further east.

We acquired a map of the major vegetationzones of the study region (Simpson 2013) fromthe Ecoshare Interagency Clearinghouse of Eco-logical Information (www.ecoshare.info/category/gis-data-vegzones; Fig. 1). Each vegeta-tion zone represents a single potential climaxvegetation type that would develop in theabsence of major disturbance and represents aunique species pool within a defined biophysicalsetting with similar climatic and topographicconditions and historical disturbance regimes(Winthers et al. 2005). Vegetation zones enablethe interpretation of structural types in a broadercontext since the same structural type mayrepresent different developmental stages in dif-ferent vegetation zones. Major vegetation zonescorrespond to those presented by Franklin andDyrness (1973) and can generally be broken intowet and dry forests. Major wet forest vegetationzones are located in the western part of theregion and include those dominated by redwood(Sequoia sempervirens) Sitka spruce (Picea sitch-


REILLY AND SPIES

ensis) and redwood (Sequoia sempervirens), Doug-las-fir (Pseudotsuga menziesii ) and tanoak (Lith-ocarpus densiflorus), western hemlock (Tsugaoccidentalis), Pacific silver fir (Abies amabilis), andmountain hemlock (Tsuga mertensiana). Major dryforest vegetation zones are located in the easternpart of the region and include those dominatedby western juniper (Juniperus occidentalis), pon-derosa pine (Pinus ponderosa), Douglas-fir (Psue-dostuga menziziii ), grand fir (Abies grandis) andwhite fir (Abies concolor), and subalpine forestsdominated by subalpine fir (Abies lasiocarpa),

Engelmann spruce (Picea engelmanii ), and white-bark pine (Pinus albicaulis).

Natural and anthropogenic disturbances influ-enced the development of forest structure in allvegetation zones. Wildfire played a major roleacross the entire study region but varied overtime (Agee 1993, Weisberg and Swanson 2003).Historic fire regimes range from high frequency,low-severity fire in warm, dry forests to lowfrequency, high-severity fire in cold and wetforests. Much of the vegetation in the region washistorically subject to a mixed-severity fire

Fig. 1. Map of major forested potential vegetation types (from Simpson 2013) in Oregon, Washington, and

inland Northern California, USA.


REILLY AND SPIES

regime where the proportion of high-severity firevaried spatially and temporally (Agee 1993, Perryet al. 2011). Several decades of fire exclusion arebelieved to have altered forest composition andstructure across much of the region, particularlyin dry forests of the eastern and southernportions of the region (Perry et al. 2011), butwildfires have increased in frequency and extentsince the mid-1980s (Littell et al. 2009). Wind-storms and landslides associated with storms offthe Pacific Ocean play a far greater role in wetforests than in dry forests, particularly in coastalareas on steep, exposed landforms (Sinton et al.2000).

In wet forests of the western portion of thestudy area, widespread clearcut logging in lowerelevation private forests began in the late 1800s,but started on federal lands in the 1940s andreached its peak in the 1970s and 1980s (Johnsonand Swanson 2009). The passage of the North-west Forest Plan (NWFP) in 1994 essentiallystopped clearcutting of older forests on federallands within the range of the Northern SpottedOwl (Strix occidentalis caurina), but clearcutting isstill the dominant practices on private forestsmanaged for timber production. Forests in thedrier eastern and southern parts of the regionwere directly affected by grazing of domesticanimals and high grade logging with theselective removal of large old-growth ponderosapine in the early to middle part of the 20thcentury (Langston 1995, Hessburg and Agee2003). Although removal of fine fuels by livestockgrazing reduced the role of fire in dry forestsduring this time, high-severity wildfires affectedsome areas in the eastern part of the region priorto the adoption of fire suppression policies thatbegan in the early 1900s (Brown 1968, Langston1995). Logging has occurred in most vegetationzones but substantial tracts of unlogged foreststill remain on federal lands (Moeur et al. 2011).

Field dataWe acquired field data on forest structure from

11,091 one ha plots from the USDA ForestService Pacific Northwest Research Station An-nual Forest Inventory and Analysis program(FIA) PNW-FIA Integrated Database (IDB). Plotswere located systematically at a density ofapproximately one every 2,400 ha in areascapable of supporting forest across all land

ownerships in Oregon, Washington, and inlandnorthern California and give an unbiased offorest conditions across the region during thestudy period. Data collection began in 2001 andcontinued until 2010 with approximately one-tenth of the plots sampled each year. Plotsincluded a series of four variable radius subplots.Live trees and snags ,12.7 cm dbh and of coverof understory vegetation (shrubs, forbs, andgraminoids) were measured in 2.1 m radiusmicroplots. Live trees and snags .12.7 cm dbhwere measured in 7.32 m radius subplots. Livetrees and snags .76.2 cm dbh on the west side ofthe Cascade Crest and .61 cm dbh on the eastside were measured in 18 m radius subplots.Dead and downed wood was sampled along two7.32 m transects in each subplot.

We acquired plot scale (one ha) summaries ofvariables describing live tree structure from theLandscape Ecology, Modeling, Mapping, andAnalysis (LEMMA) Project (http://lemma.forestry.oregonstate.edu). These variables includ-ed the following: basal area (BA), density (TPH),live tree biomass (BPH), quadratic mean diame-ter of all dominant and codominant trees (QMD),basal area-weighted mean dbh of all live trees(DBH), a diameter diversity index (DDI) basedon the tree densities in different size classes(Spies et al. 2007), standard deviation of dbh(SDDBH), average height of all dominant andcodominant trees (STNDHT), and percent cano-py cover of all live trees (CC) as calculated usingmethods in the Forest Vegetation Simulator (FVS)(Crookston and Stage 1999). This method cor-rects for overlapping canopies and canopyclosure occurs at about 65%. Variables describingdead wood included the density (SDPH), andbiomass (SBPH) of snags .12 cm dbh and .2 mtall, as well as the biomass of dead and downedwood .12 cm at the large end and .3 m long(BDW). The only non-tree variable included totalcover of understory vegetation (USC) which wascalculated as the summed total percent cover ofshrubs, forbs, and graminoids.

We acquired spatial data on the perimeters ofwildfires greater than 400 ha from 1984 to 2010from the Monitoring Trends in Burn SeverityProgram (www.mtbs.gov) and overlaid plotlocations to classify plots as burned or unburned.


REILLY AND SPIES

Data analysisWe used principal components analysis (PCA)

to reduce the dimensionality of the previouslylisted metrics of stand structure since manystructural attributes are linearly related (e.g., livetree biomass and basal area). We used theprincomp function in R with the cross-productsmatrix based on the correlation matrix amongvariables. All variables were centered to a meanof zero and standardized in order to equalizevariance among variables measured in differentunits.

We performed hierarchical agglomerative clus-tering with Euclidean distance and Ward’smethod (Ward 1963) to assign each plot intostructure-based clusters. Agglomerative cluster-ing works by iteratively merging the most similarplots into groups until all groups are merged. Weused the hclust.vector algorithm in the packagefastcluster in R (R Development Core Team2011). Plots were clustered on the axis scores ofthe first five dimensions of the PCA to reduceredundancy of correlated structural variables(e.g., basal area and biomass) as well assubjectivity or bias in selection of specificvariables.

We used data on ages of dominant and co-dominant trees to assign an estimated stand agefor each plot. Age of individual trees wasestimated in the field from increment cores takenfrom one live dominant and co-dominant tree foreach species. In cases where trees were too largefor the increment borer to reach the pith of thetree, age was estimated based on growth in theinner five cm of the core. Stand age was thenestimated as the basal area weighted average age

of all dominant and co-dominant tree ages in aplot. We compared the plot level median basalarea weighted ages of dominant and co-domi-nants first among structural classes and thenamong vegetation zones by individual structuralclasses using notched boxplots. Notched box-plots show the median, interquartile range, andnotches that approximate a 95% confidenceinterval around median values where the lackof overlap between the notches of two boxesprovides strong evidence that medians differ(Chambers et al. 1983).

RESULTS

Principal components analysisof structural variation

The first five PCA components accounted for90% of the variance explained (Table 1). Compo-nent 1 was strongly and negatively correlatedwith live tree biomass and basal area. Compo-nent 2 had a strong positive correlation with thedensity of trees per ha and strong negativecorrelation with quadratic mean diameter. Com-ponent 3 had a strong positive correlation withthe biomass and density of snags and biomass ofdead and downed wood, but a strong negativecorrelation with density of live trees. Component4 was negatively correlated with the cover ofunderstory vegetation and biomass of dead anddowned wood. Component 5 was negativelywith the biomass of dead and downed wood, butwas also strongly and positively correlated withthe density of snags per ha and the cover ofunderstory vegetation.

Table 1. Proportion of variance explained for the first five dimensions of a principal components analysis on

attributes of forest structure in Oregon, Washington, and inland Northern California, USA.

Attribute Axis 1 Axis 2 Axis 3 Axis 4 Axis 5

Proportion of variance explained (total ¼ 0.90) 0.47 0.16 0.13 0.09 0.06Basal area �0.39 . . . . . . . . . . . .Live tree density . . . 0.60 �0.33 . . . . . .Live tree biomass �0.40 . . . . . . . . . . . .Quadratic mean diameter �0.25 �0.53 . . . . . . . . .Diameter diversity index �0.38 . . . . . . . . . . . .Standard deviation of diameter at breast height �0.34 �0.27 . . . . . . . . .Stand height �0.35 �0.24 . . . . . . . . .Canopy cover �0.33 0.29 �0.3 . . . . . .Snag density . . . 0.24 0.56 . . . 0.42Snag biomass �0.23 . . . 0.58 . . . . . .Biomass of dead and downed wood . . . . . . 0.32 �0.34 �0.81Understory cover . . . . . . . . . �0.92 0.35


REILLY AND SPIES

Classification of structural typesWe began interpreting clusters from the

classification by iteratively ‘‘cutting’’ the dendro-gram (Fig. 2) at increasing numbers of clustersrepresenting broad but ecologically interpretablestructural groupings with the goal of represent-ing variation along all five PCA componentswhile still maintaining a manageable number ofclusters. The hierarchy of clusters begins with‘‘groups’’ at the coarsest level which are thenbroken into ‘‘classes,’’ and then finally into‘‘types.’’ The initial cut resulted in three clustersrepresenting low (,25 Mg/ha), moderate (25–99Mg/ha), and high live tree biomass (.100 Mg/ha)structural groups. The second cut resulted ineleven clusters that broke groups into ‘‘structuralclasses’’ that differed in live tree density andQMD (Fig. 3). We continued identifying clustersto include structural variation along the othercomponents including snags, understory cover,and biomass of dead and downed wood (Table2). Structural types also varied in relativeabundance among vegetation zones (Table 3),geographic distributions (Appendix A), size classdistributions of live trees, snags, dead anddowned wood, and the proportion of understorycover composed of shrubs, forbs and graminoids(Appendix B).

Age differences among structural classesMedian age of structural classes varied among

vegetation zones (Figs. 4 and 5). Age generallyincreased with live biomass in wet vegetationzones, but showed little relationship to biomassin dry vegetation zones, with the exception ofDouglas-fir and grand fir/white fir where medianage was greatest in the highest biomass classes.Median age of dominant and co-dominant treesfrequently differed among vegetation zones forindividual structural classes (Appendix C). Themajor exceptions to this were in dry vegetationzones and in Structural Class 1 (Very Low LiveBiomass and Density) where the median age waszero and there was no evidence of differencesamong vegetation zones. Median age of struc-tural classes was generally lowest in westernhemlock and Sitka spruce-redwood vegetationzones and highest in mountain hemlock andwestern juniper zones.

DISCUSSION

This study is the first empirically basedcharacterization of multiple components of foreststructure for any region and reveals the com-plexity of forest structure across vegetation zonesand environments. Although most variation inforest structure was related to live tree biomass,basal area, and the presence of big trees, other

Fig. 2. Dendrogram with hierarchical classification of forest structural groups, classes, and types in Oregon,

Washington, and inland Northern California.


REILLY AND SPIES

components related to size and density of trees,the abundance of snags and dead and downedwood, and understory vegetation accounted foralmost as much additional variation. Understoryvegetation acted independently while snags anddead and downed wood emerged at multiplepoints during biomass development suggestingthat the timing and importance of variousecological processes (e.g., mortality, growth,recruitment) likely varies during different devel-opmental stages in different vegetation zones.These results indicate that conceptual models ofstand development based on a monotonic trajec-tory of biomass accumulation following stand-replacing disturbance are inadequate to accountfor the diversity of ways that interactions amongvegetation, environment, and disturbance maybe expressed through combinations of individual

structural components.

Structural pattern and ecological processesOur study of forest structure is novel in that it

provides an unbiased sample of the variety ofstructural conditions across 25 million ha offorest as they ‘‘existed’’ during the study periodwithout having to fit observations into precon-ceived classes. The traditional approach to thestudy of structural development is based onobservations at a few subjectively selected sites tocharacterize conditions that fit preconceivedstages identified by tree size or age (e.g., ‘‘old-growth or ‘‘early-seral’’; e.g., Spies and Franklin1991). This traditional approach has the ability tocontrol for environment and disturbance historywhile making inferences about the roles ofecological processes during structural develop-

Fig. 3. Illustrations of structural classes and selected structural types in Oregon, Washington, and inland

Northern California in relation to live tree biomass and density. Levels of snag density, biomass of dead and

downed wood, and understory vegetation are indicated in boxes along the borders of each cell.


REILLY AND SPIES

ment, but is based on a limited range ofstructural variation and disturbance history,potentially biasing findings and missing struc-tural conditions that don’t fit into the simplestand development models. The strength of ourapproach is the ability to characterize the fullrange of structural conditions as it exists whetheror not the types fit into preconceived models. Weview the two approaches as complimentary. Anunbiased regional sample of the diversity andabundance of different structural conditions candetect structural types that arise from bothcommon and relatively rare (limited by theintensity of the sample) ecological patterns andprocesses including different environment anddisturbance regimes. An unbiased characteriza-tion of forest structure can be used to generatehypotheses on putative processes inferred from

structural pattern that more traditional, intensivelocalized studies can test with better control forenvironment and disturbance history.

Patterns in ecological systems are often theresult of multiple processes, and any predictionregarding pattern must be derived not only fromthe patterns themselves, but from an understand-ing of the variety of processes that might beoperating to create them (Cale et al. 1989). Pastconceptual models of structural developmenthave embraced pattern and process and de-scribed similar stages based on the dominantprocesses and resulting structural conditions thatoccur during a particular range of stand ages(e.g., Bormann and Likens 1979, Oliver andLarson 1990, Franklin et al. 2002). Many of ourstructural classes and types correspond well withpreviously described stages (Fig. 6) and provide

Table 2. Mean and standard deviation of nine stand level structural attributes� used in the hierarchical cluster

analysis for eleven forest structural classes and twenty-five structural types in Oregon, Washington, and inland

Northern California, USA.

Class/type

BPH(Mg/ha) BA (m2/ha) CC (%) QMD (cm) TPH (/ha) SDPH (/ha)

SBPH(Mg/ha)

BDW(Mg/ha) USC (%)

1 0 6 0.1 0 6 0.1 0 6 3 0.2 6 0.9 19 6 116 129 6 205 24 6 50 19 6 21 45 6 351.1 0 6 0.2 0.1 6 0.2 1 6 5 0.4 6 1.4 47 6 186 343 6 225 64 6 70 20 6 26 40 6 331.2 0 6 0 0 6 0 0 6 0 0 6 0.3 4 6 44 16 6 29 3 6 7 19 6 18 47 6 362 8 6 7 4 6 3 13 6 9 23 6 13 166 6 164 8 6 30 1 6 4 4 6 10 48 6 302.1 10 6 6 6 6 3 15 6 9 26 6 13 175 6 154 3 6 8 0.3 6 1 2 6 5 46 6 282.2 2 6 3 1 6 1 8 6 6 16 6 9 146 6 183 19 6 49 2 6 6 8 6 16 54 6 343 16 6 14 9 6 5 45 6 17 11 6 4 923 6 538 13 6 20 2 6 6 28 6 34 84 6 364 42 6 25 14 6 7 30 6 13 37 6 16 195 6 145 7 6 13 1 6 3 7 6 13 55 6 355 42 6 38 17 6 10 45 6 20 19 6 9 784 6 596 21 6 38 3 6 8 8 6 12 18 6 205.1 27 6 21 14 6 8 47 6 18 13 6 4 1115 6 626 2 6 8 0.4 6 2 6 6 8 26 6 195.2 25 6 11 12 6 4 30 6 12 23 6 9 389 6 306 22 6 25 2 6 4 7 6 8 24 6 215.3 79 6 48 26 6 11 61 6 18 21 6 8 923 6 557 42 6 57 8 6 13 13 6 18 0.3 6 16 69 6 41 21 6 9 42 6 16 32 6 14 359 6 287 115 6 156 23 6 37 13 6 14 47 6 316.1 73 6 40 22 6 9 45 6 14 31 6 11 363 6 222 73 6 64 12 6 11 12 6 11 47 6 306.2 43 6 36 11 6 8 23 6 17 42 6 27 329 6 563 413 6 251 96 6 62 25 6 23 45 6 387 69 6 38 27 6 11 70 6 16 16 6 6 1926 6 1528 19 6 23 2 6 4 10 6 11 55 6 347.1 70 6 29 25 6 8 62 6 14 20 6 4 838 6 305 15 6 21 2 6 3 7 6 7 57 6 347.2 86 6 41 31 6 10 76 6 13 14 6 3 2060 6 777 27 6 23 3 6 4 13 6 12 53 6 307.3 46 6 37 23 6 13 78 6 14 9 6 2 3701 6 1849 15 6 23 2 6 5 13 6 13 55 6 408 157 6 80 37 6 13 69 6 15 32 6 11 584 6 374 28 6 34 5 6 11 29 6 36 83 6 408.1 123 6 47 33 6 10 67 6 14 28 6 7 624 6 316 22 6 21 3 6 3 10 6 9 89 6 408.2 139 6 68 34 6 12 69 6 16 29 6 9 684 6 467 39 6 46 9 6 16 65 6 42 68 6 388.3 234 6 83 46 6 15 71 6 14 43 6 12 399 6 243 26 6 28 5 6 8 17 6 18 92 6 389 157 6 82 44 6 16 80 6 13 20 6 6 1757 6 1138 79 6 73 13 6 14 21 6 26 39 6 329.1 98 6 50 32 6 11 71 6 14 18 6 6 1555 6 1005 97 6 50 14 6 11 20 6 16 57 6 309.2 194 6 79 51 6 13 82 6 11 23 6 6 1527 6 907 44 6 34 7 6 7 10 6 9 28 6 259.3 163 6 72 47 6 15 87 6 10 16 6 4 2681 6 1388 134 6 120 26 6 22 49 6 42 36 6 3610 262 6 130 55 6 19 77 6 13 33 6 12 820 6 515 120 6 98 33 6 32 29 6 25 29 6 3310.1 206 6 97 46 6 16 72 6 15 31 6 10 769 6 514 163 6 111 38 6 27 31 6 21 61 6 3310.2 266 6 123 56 6 18 79 6 12 33 6 12 834 6 497 74 6 64 19 6 17 25 6 19 10 6 1210.3 383 6 131 71 6 18 85 6 9 36 6 11 900 6 558 155 6 90 62 6 47 39 6 43 12 6 1611 482 6 165 75 6 22 83 6 12 46 6 18 647 6 537 73 6 57 40 6 37 67 6 57 73 6 3911.1 483 6 163 76 6 23 85 6 13 41 6 15 825 6 620 87 6 63 53 6 41 90 6 63 57 6 3111.2 482 6 169 73 6 20 80 6 9 54 6 18 406 6 240 54 6 38 21 6 18 35 6 26 94 6 39

� BPH¼ Live Tree Biomass, BA¼ Basal Area, CC¼Canopy Cover, TPH¼ Live Tree Density, SDPH¼ Snag Density, SBPH¼Biomass of Snags, BDW¼ Biomass of Dead and Downed Wood, and USC¼ Understory Cover.


REILLY AND SPIES

some support for these models in forests where

stand-replacing disturbance has been common

historically and biomass tends to accumulate

monotonically with age. However, broad overlap

in age among structural classes in dry vegetation

zones indicates that ‘‘stand’’ age is a poor

surrogate for structure in forests where distur-

bance regimes include low-severity or partial

stand-replacing disturbance and age structure is

far more complex (Muir 1993, Taylor and Skinner

2003, Taylor 2010).

Differences in productivity, the range of

variation in individual components of structure

(e.g., live biomass), and the timing of several

ecological processes among different vegetation

zones limit the generalizability of process- and

age-based conceptual models of structural devel-

opment at a regional scale. Our structure-based

classification offers an alternative framework for

conceptualizing structural development, but

structure alone may be equivocal since many

structural types potentially developed through

multiple pathways and often represent different

stages of development in any given vegetation

zones. An age- or process-based understanding is

evidently still necessary to provide an ecological

interpretation of structural patterns in terms of

development (e.g., early vs. old-growth; Tables

4–6; Appendix D). These are inevitably contin-

gent on subjectively binning a continuous pro-

cess into a discreet stage-based framework with

some implicit temporal progression, but can still

be applied with explicit recognition that devel-

opment may proceed in a multi-directional, non-

linear fashion.

Most structural classes and types in the low

biomass group correspond well with previously

described early developmental stages (Fig. 6)

Table 3. Relative abundance (%) of eleven major structural classes and twenty-five structural types by vegetation

zone� (Simpson 2013) in Oregon, Washington, and inland Northern California, USA.

Class type SS-RW DF-TO WH SF MH WJ PP DF GF/WF SA All

1 1.7 2.2 2.4 0.8 0.9 1.1 2.3 2.2 2.1 7.0 2.51.1 0.0 1.9 0.1 0.1 0.9 0.0 0.4 1.1 0.9 5.8 0.81.2 1.7 0.3 2.3 0.6 0.0 1.1 1.9 1.1 1.2 1.2 1.62 0.3 2.2 1.9 1.3 1.3 54.6 15.3 5.3 4.3 2.3 7.42.1 0.3 1.3 0.4 0.5 1.1 41.6 12.2 3.0 2.9 1.7 5.02.2 0.0 0.9 1.5 0.8 0.2 13.0 3.2 2.3 1.5 0.6 2.33 5.5 1.6 7.5 5.7 1.6 0.0 2.9 6.7 3.2 3.2 4.94 1.7 0.9 2.6 1.3 1.3 18.6 21.6 11.0 9.9 4.4 8.95 5.9 4.1 5.8 3.3 5.9 21.2 27.3 12.0 12.5 7.6 11.05.1 2.1 0.3 2.6 1.4 1.1 8.9 11.4 2.8 3.3 2.3 3.65.2 0.7 0.6 0.4 0.5 1.4 11.9 12.9 4.5 5.2 2.3 4.15.3 3.1 3.1 2.8 1.4 3.4 0.4 3.0 4.6 4.0 2.9 3.26 2.4 2.5 2.7 0.5 4.5 2.6 8.7 7.7 11.3 14.0 6.36.1 2.4 1.3 2.5 0.5 3.2 2.6 8.6 7.2 9.7 8.2 5.56.2 0.0 1.3 0.1 0.0 1.3 0.0 0.1 0.5 1.6 5.8 0.87 15.9 19.7 12.2 14.4 12.7 1.5 15.7 20.4 14.2 20.7 14.87.1 4.8 3.8 5.4 5.1 2.9 0.7 8.7 9.8 6.6 6.4 6.47.2 6.6 7.8 3.7 4.6 6.5 0.0 3.6 6.4 5.3 7.3 4.87.3 4.5 8.1 3.1 4.7 3.4 0.7 3.3 4.2 2.3 7.0 3.68 26.9 8.4 25.9 12.4 5.4 0.4 2.3 10.8 10.2 7.6 12.58.1 10.0 1.9 8.3 4.6 2.5 0.4 1.9 5.8 5.1 2.9 5.28.2 9.7 5.0 10.1 5.4 1.4 0.0 0.4 1.8 3.1 4.4 4.08.3 7.2 1.6 7.6 2.4 1.4 0.0 0.0 3.2 2.0 0.3 3.39 13.4 35.6 13.5 16.7 31.9 0.0 3.6 14.0 14.8 19.2 14.29.1 4.8 4.1 3.8 4.2 9.3 0.0 1.6 4.2 5.9 13.4 4.79.2 5.2 25.6 6.3 7.4 12.9 0.0 1.9 8.9 6.6 1.5 6.89.3 3.4 5.9 3.4 5.1 9.7 0.0 0.1 1.0 2.3 4.4 2.710 15.5 19.7 13.4 20.9 29.0 0.0 0.1 8.0 15.2 14.0 12.110.1 4.1 2.5 5.7 7.0 10.8 0.0 0.0 2.0 6.0 10.8 4.410.2 8.3 14.7 5.6 6.3 11.8 0.0 0.1 5.6 7.6 3.2 5.710.3 3.1 2.5 2.1 7.7 6.5 0.0 0.0 0.5 1.7 0.0 1.911 10.7 3.1 12.1 22.7 5.6 0.0 0.0 1.9 2.1 0.0 5.511.1 4.5 1.6 6.2 17.1 4.1 0.0 0.0 0.7 1.1 0.0 3.211.2 6.2 1.6 5.9 5.6 1.4 0.0 0.0 1.2 1.0 0.0 2.3

� WJ ¼ Western Juniper, PP ¼ Ponderosa Pine, GF/WF ¼ Grand Fir/White Fir, DF ¼ Douglas-fir, SS-RW ¼ Sitka Spruce–Redwood, DF-TO¼ Douglas-fir–Tanoak, WH¼Western Hemlock, SF ¼ Silver Fir, and MH¼Mountain Hemlock.


REILLY AND SPIES

Fig. 4. Boxplots of mean age of all dominant and co-

dominant trees by structural class for wet vegetation

zones in Oregon, Washington, and inland Northern

California: Sitka-spruce/redwood, Douglas-fir tanoak,

western hemlock, silver fir, and mountain hemlock.

The width of each boxplot is proportional to the square

root of the sample size and means are represented by

an asterisk.

Fig. 5. Boxplots of mean age of all dominant and co-

dominant trees by structural class for dry vegetation

zones in Oregon, Washington, and inland Northern

California: western juniper, ponderosa pine, Douglas-

fir, grand fir/white fir, and subalpine forests. The width

of each boxplot is proportional to the square root of the

sample size and means are represented by an asterisk.


REILLY AND SPIES

Fig. 6. Correspondence of traditional conceptual models of strucural development with empirically based

structural classes in this study.

Table 4. Characterization of developmental stage, potential developmental pathways, and supporting lines of

evidence� for low biomass structural types in Oregon, Washington, and inland Northern California, USA, by

vegetation zone� (Simpson 2013).

Structural typeDevelopmental

stagePotential developmental

pathways Vegetation zone Evidence

(1.1) Very Low Live Biomass andDensity w/ Very High SnagDensity

early stand-replacing fire and/orepidemic pathogen

all AGE, RFH

(1.2) Very Low Live Biomass andDensity w/out Snags

early recent clearcutting, salvagefollowing recent fire and/orepidemic pathogen, forestexpansion

all AGE, RFH,CMP

(2.1) Low Live Biomass andDensity w/ Medium Trees, (2.2)Low Biomass and Low Density

early recent clearcutting, forestexpansion

SS-RW, DF-TO, WH, SF,MH, PP, GF/WF, DF, SA

AGE, PROD,CMP

mid forest expansion WJ AGE, PROD

(3.0) Low Live Biomass and HighDensity

early forest expansion SS-RW, DF-TO, WH, SF,MH, PP, GF/WF, DF, SA

AGE, PROD

clearcutting AGE, HIST

� Lines of evidence include stand age (AGE), recent fire history since 1984 (RFH), productivity (PROD), disturbance andmanagement history (HIST), and current management practices (CMP).

� WJ ¼ Western Juniper, PP ¼ Ponderosa Pine, GF/WF ¼ Grand Fir/White Fir, DF ¼ Douglas-fir, SS-RW ¼ Sitka Spruce–Redwood, DF-TO¼ Douglas-fir–Tanoak, WH¼Western Hemlock, SF¼ Silver Fir, and MH ¼Mountain Hemlock.


REILLY AND SPIES

Table 5. Characterization of developmental stage, potential developmental pathways, and supporting lines of

evidence� for moderate biomass structural types in Oregon, Washington, and inland Northern California,

USA, by vegetation zone� (Simpson 2013).

Structural type Stage Potential developmental pathways Vegetation zone Evidence

(4.0) Moderate LiveBiomass and LowDensity w/ Big Trees

mid variable retention harvest SS-RW, DF-TO,WH

CMP

forest expansion SF AGE, PRODmature forest expansion, low-severity fire MH, PP, GF/WF,

DF, SAlate forest expansion WJ

(5.1) Moderate LiveBiomass and Very HighDensity

mid mid/late 20th century clearcutting SS-RW, DF-TO,WH

AGE, HIST

forest expansion SF, MH, SA AGEinfilling from fire exclusion and high grading,early 20th century stand-replacing fire

GF/WF, DF AGE, HIST

infilling from fire exclusion and high grading PPmature forest expansion, infilling WJ

(5.2) Moderate LiveBiomass and Density

mid mid/late 20th century clearcutting andthinning

SS-RW, DF-TO,WH

AGE, HIST

late 19th/early 20th century stand-replacingfire, forest expansion

SF, MH, SA

infilling from fire exclusion and high grading,early 20th century stand-replacing fire,recent thinning or prescribed fire

GF/WF, DF AGE, HIST,CMP

infilling from fire exclusion and high grading,recent thinning or prescribed fire

PP

mature forest expansion, infilling WJ AGE

(5.3) Moderate LiveBiomass and HighDensity w/ Big Trees


AGE


SF

mature late 19th/early 20th century stand-replacingfire, forest expansion

MH, SA

infilling from fire exclusion and high grading,early 20th century stand-replacing fire

GF/WF, DF AGE, HIST

infilling from fire exclusion and high grading PPold-growth forest expansion, infilling WJ

(6.1) Moderate LiveBiomass and Density w/Big Trees and HighSnag Density

mature recent low-severity fire, endemic insect and/orpathogen

SS-RW, DF-TO,WH, SF

AGE, RFH

late MH, SA, GF/WF, DF, PP

old-growth WJ

(6.2) Moderate LiveBiomass and Density w/Big Trees and Very HighSnag Density

early recent partial stand-replacing fire and/orepidemic insect, pathogen

ALL RFH

(7.1) Moderate LiveBiomass and HighDensity, (7.2) ModerateLive Biomass and VeryHigh Density


AGE, HIST


SF

mature late 19th/early 20th century stand-replacingfire, forest expansion

MH, SA


GF/WF, DF

infilling from fire exclusion and high grading PP


REILLY AND SPIES

and have characteristics of early-seral habitatsincluding sparse canopy cover and low basalarea (Swanson et al. 2010). Variation in thedensity of live trees, snags, and dead anddowned wood are indicative not only of thepotential diversity of structural conditions inearly developmental stages (Fig. 3) depending onrecent disturbance history (Table 4), but alsovariation in the role of different ecologicalprocesses during early developmental stagesdepending on productivity. Snags and dead anddowned wood may be abundant followinglegacy creation in wildfires in Type 1.1 (90%burned since 1984), but are absent during standinitiation and cohort establishment followinglogging or expansion into previously non-forest-ed areas (Type 1.2). Differences in live treedensity, canopy cover, and age between Struc-tural Classes 2 (Low Live Biomass and Density)and 3 (Low Live Biomass and High Density)suggest variation in the roles of several ecologicalprocesses and a major dichotomy in earlydevelopmental pathways. Canopy closure mayoccur relatively rapidly as in Structural Class 3where rapid growth and continuous recruitmentoffset high levels of density-dependent mortalityin smaller, less competitive trees (Lutz andHalpern 2006). These processes are likely lessimportant in Structural Class 2 where trees areolder and density and canopy cover are muchlower. This stage represents a period of protract-ed recruitment either not described or lessemphasized in traditional models of structuraldevelopment. Cohort establishment may beepisodic from recruitment limitation related toclimatic fluctuations (Brown and Wu 2005, Zaldet al. 2012), competition from shrubs or herba-

ceous vegetation (Putz and Canham 1992, Rigi-nos 2009), frequent fire (Taylor 2010), dispersallimitation (Agee and Smith 1984), or a combina-tion of these factors (Acacio et al. 2007). Whilecompetition may play an important role inrapidly regenerating stands at high density,facilitation may be more important to overcomerecruitment limitation in harsh environments(Berkowitz et al. 1995, Callaway 1998, Calderand St. Clair 2012, Rice et al. 2012).

Structural types in the moderate live biomassgroup generally lack dead wood and represent awide array mid, mature, and late developmentalstages that have resulted from a variety ofdevelopmental pathways in different vegetationzones (Table 5). Structural Classes 5 (ModerateLive Biomass and High Density) and 7 (ModerateLive Biomass and Very High Density) resemblepreviously described stages (Fig. 6) that arecharacterized by rapid growth and biomassaccumulation following stand-replacing distur-bance (Franklin et al. 2002). Tremendous varia-tion in density among the types that make upthese classes reflect differences in productivityand management practices in the vegetationzones where they occur. Structural Classes 4(Moderate Live Biomass and Low Density withBig Trees) and 6 (Moderate Live Biomass andDensity with Big Trees and Snags) resemblewoodlands and savannahs with sparse canopyand big trees. These lack analogs in currentmodels of stand development, but are recognizedas important alternative stable states across muchof the world (Staver et al. 2011). The geographicdistribution of Structural Class 4 across the drier,hotter eastern part of the study region (Appen-dix: Fig. A2) suggests slow growth and low

Table 5. Continued.

Structural type Stage Potential developmental pathways Vegetation zone Evidence

(7.3) Moderate LiveBiomass and ExtremelyHigh Density


AGE, HIST


SF, MH, SA AGE, PROD


GF/WF, DF AGE, HIST

infilling from fire exclusion and high grading PPold-growth forest expansion, infilling WJ AGE, HIST


� WJ ¼Western Juniper, PP ¼ Ponderosa Pine, GF/WF ¼ Grand Fir/White Fir, DF ¼ Douglas-fir, SS-RW ¼ Sitka Spruce–Redwood, DF-TO¼ Douglas-fir–Tanoak, WH¼Western Hemlock, SF ¼ Silver Fir, and MH¼Mountain Hemlock.


REILLY AND SPIES

Table 6. Characterization of developmental status, potential developmental pathways, and supporting lines of

evidence� for high biomass structural types in Oregon, Washington, and inland Northern California, USA, by

vegetation zone� (Simpson 2013).

Structural type StagePotential

developmental pathways Vegetation zone Evidence

(8.1) High Live Biomass andDensity w/ Big Trees and LowDead Biomass

mature early/mid 20th clearcutting SS-RW, DF-TO,WH

AGE, HIST

late 19th century stand-replacing fire, forestexpansion

SF

late late 19th century stand-replacing fire, forestexpansion

MH, SA

frequent pre-20th century fire, infillingfrom fire exclusion and high grading,late 19th century stand-replacing fire

GF/WF, DF

old-growth frequent pre-20th century fire, infillingfrom fire exclusion

PP

(8.2) High Biomass and Densityw/ Big Trees and Very HighBiomass of Dead and DownedWood

mature early/mid 20th century stand-replacing fire SS-RW, DF-TO,WH

AGE, HIST

late 19th century stand-replacing fire SFlate late 19th century stand-replacing fire MH, SA AGE, PROD

frequent pre-20th century fire, infillingfrom fire exclusion and high grading,late 19th century stand-replacing fire

GF/WF, DF AGE, HIST


PP

(8.3) High Live Biomass andModerate Density w/ Big Trees

late mid 20th century clearcutting, recentthinning

SS-RW, DF-TO,WH

AGE, HIST,CMP

late 19th century stand-replacing fire, forestexpansion

SF, MH, SA AGE, HIST

old-growth frequent pre-20th century fire, late 19thcentury stand-replacing fire

GF/WF, DF

(9.1) High Live Biomass andVery High Density w/ BigTrees and Moderate DeadBiomass

mature early 20th century stand-replacing fire SS-RW, DF-TO,WH

AGE, HIST

19th century stand-replacing fire, forestexpansion

SF, MH, SA

infilling from fire exclusion and highgrading, late 19th century stand-replacing fire

GF/WF, DF

late frequent pre-20th century fire, infillingfrom fire exclusion and high grading

PP

(9.2) High Live Biomass andVery High Density w/ BigTrees and Low Dead Biomass

late early 20th century stand-replacing fire SS-RW, DF-TO,WH


SF

old-growth 19th century stand-replacing fire, forestexpansion

MH, SA

frequent pre-20th century fire, infillingfrom fire exclusion

GF/WF, DF, PP

(9.3) High Live Biomass andExtremely High Density w/ BigTrees and High Dead Biomass

late early 20th century stand-replacing fire SS-RW, DF-TO,WH


SF, MH, SA

frequent pre-20th century fire, infillingfrom fire exclusion

GF/WF, DF


PP


REILLY AND SPIES

recruitment while lack of snags and low densityindicate very little competition related mortality.In contrast, Structural Class 6 represents analternative stage where very high but variablesnag density reveals a gradient in recent levels ofmortality suggestive of low- or moderate-sever-ity partial stand-replacing disturbances (10%burned since 1984) which decrease live treebiomass and density, but increase the dominanceof bigger trees. Structural Type 6.2 (ModerateLive Biomass and Density with Big Trees andVery High Snag Density and Biomass) is em-blematic of the diverse structural conditionscreated by wildfire (70% burned since 1984) inlater stages of development where elements ofboth early (e.g., high snag density, sparse canopycover) and later developmental stages (e.g., big,old remnant trees, big snags, abundant deadwood) can be found.

Structural classes in the high live biomassgroup generally resemble later developmentalstages (Table 6) described in previous models ofstand development (Fig. 6) and corroborate thediversity of structural conditions that can befound in mature and old-growth stages (Franklinet al. 2002). Mortality during these stages ishypothesized to shift towards density-indepen-

dent sources such as wind, insects, and disease(Franklin et al. 2002), but results supporting thishave been mixed (see Lutz et al. 2014). Largedifferences in the density of live trees and smallsnags among high biomass structural classes(Appendix B) indicate that the role of density-dependent mortality may vary considerablyduring later developmental stages, especially indifferent vegetation zones where density in laterstages may also be variable (Table 2). Very highdensity of small trees and snags in StructuralClass 9 (High Live Biomass and Very HighDensity with Big Trees) indicate that density-dependent mortality can be very high duringunderstory re-initiation or vertical diversifica-tion. On the contrary, low density of small snagsand trees in Structural Class 8 (High LiveBiomass and Density with Big Trees) indicatelittle density-dependent mortality where under-story re-initiation has been either been delayedby dispersal limitation (Wimberly and Spies2001) or prevented by dense understory vegeta-tion (George and Bazazz 1999), browsing (Roo-ney and Waller 2003), or frequent low-severityfire (Platt et al. 1988). Although high levels ofdensity-dependent mortality may operate at finescales where aggregated recruitment occurs in

Table 6. Continued.

Structural type StagePotential

developmental pathways Vegetation zone Evidence

(10.1) Very High Live Biomassand High Density w/ LargeTrees and High Dead Biomassand Understory Cover

late centuries since stand-replacing disturbance SS-RW, DF-TO,WH, SF

old-growth MH, SA, GF/WF, DF

(10.2) Very High Live Biomassand High Density w/ LargeTrees and Moderate DeadBiomass

late SS-RW, DF-TO,WH, SF

old-growth MH, GF/WF,DF, SA, PP

(10.3) Extremely High LiveBiomass w/ Large Trees andHigh Dead Biomass

MH, GF/WF,DF, SA

(11.1) Extremely High LiveBiomass w/ Giant Trees andHigh Density, (11.2) ExtremelyHigh Live Biomass w/ GiantTrees and Moderate Density

SS-RW, DF-TO,WH, SF, MH,GF/WF, DF


� WJ ¼Western Juniper, PP ¼ Ponderosa Pine, GF/WF ¼ Grand Fir/White Fir, DF ¼ Douglas-fir, SS-RW ¼ Sitka Spruce–Redwood, DF-TO¼ Douglas-fir–Tanoak, WH¼Western Hemlock, SF ¼ Silver Fir, and MH¼Mountain Hemlock.


REILLY AND SPIES

gaps, it may have little effect on structural

change (Lutz et al. 2014). Instead, low levels of

density-independent mortality of big trees from

windthrow and endemic infestations of insects

and disease (Franklin et al. 2002) that increase

biomass of snags and dead and downed wood

have greater implications for structural develop-

ment in Structural Classes 10 (Very High Live

Biomass with Large Trees) and 11 (Extremely

High Live Biomass with Giant Trees).

Potential pathways of structural development

While many of our structural classes fit into

some stages of existing models of stand devel-

opment, it is clear that there is a great deal of

variation in structure within any of these existing

stages of development. It is also clear that the

pathways of development and expression of

different structural types varies greatly across

the region. The relative abundance of structural

classes and types in individual vegetation zones

corresponds with major gradients in climate and

productivity but also reflects legacies of historical

and contemporary management practices and

disturbance regimes. Together, productivity and

disturbance regimes interact to constrain the

abundance and diversity of structural conditions

and developmental pathways in an individual

vegetation zone. Below, we present conceptual

models of four of the major types of structural

development that we hypothesize have given rise

to the some structural classes and types observed

Fig. 7. Conceptual models of potential pathways of structural development in western hemlock, mountain

hemlock, western juniper, and grand fir/white fir vegetation zones in Oregon, Washington, and inland Northern

California.


REILLY AND SPIES

in the region (Fig. 7). We limit discussion to a fewdominant pathways supported by the abundanceof structural classes or types by vegetation zoneand the available literature. The proposed path-way types are distinguished based on the rangeof variation along various components of struc-ture (e.g., biomass and density), the rate andtiming of transitions among stages, the relativeimportance of different disturbance types, andcomplexity of pathways.

Structural classes and potential pathways ofdevelopment in the western hemlock zonecorrespond with existing models of developmentfor infrequent, stand-replacing disturbance re-gimes (e.g. Douglas-fir/western hemlock, Frank-lin et al. 2002; Fig. 7). Regeneration, live biomass,and density may develop rapidly during earlystand development (Lutz and Halpern 2006). Thedead wood component can reach its peakfollowing fire (Spies et al. 1988), but was mostabundant in later developmental stages aslogging and infrequent fire during the 20thcentury (Curtis et al. 1998, Healey et al. 2008,Johnson and Swanson 2009,) have limited theoccurrence of diverse early-seral conditions withlive and dead legacies which made up a verysmall proportion of this vegetation zone (;0.2%).Although protracted establishment (;40 years toend of establishment stages) was common in thepast (Winter et al. 2002, Tepley et al. 2014),management practices and current policy (Ore-gon Revised Statutes 527.745; http://www.oregonlaws.org/ors/527.745) have reduced theduration of early development stages (Hansenet al. 1991). Application of silvicultural treat-ments like pre-commercial thinning, thinning(Chan et al. 2006), or variable retention harvest(Aubry et al. 2009) may alter developmentaltrajectories by decreasing density, increasingdominance of big trees, and accelerating thedevelopment of structural development (Baileyand Tappeiner 1998). Low- and moderate-sever-ity wildfire played an historic role during laterdevelopmental stages in some parts of thewestern hemlock zone (Tepley et al. 2013), butthese have affected only a small proportion of thevegetation zone during the 20th century. Currentdynamics in later stages may be driven primarilyby stand-level processes rather than externaldisturbance agents (Lutz et al. 2014). Multipleold-growth structural conditions and stages are

possible as live biomass may fluctuate frommortality of individual large trees or loss offounding cohorts, increasing dead and downedwood and density through gap phase recruit-ment (Gray and Spies 1996).

In contrast to the western hemlock zone, forestmanagement has been minimal in the mountainhemlock zone and structural types and pathwaysdevelopment are primarily the result of naturalprocesses (Fig. 7). Live biomass and density riseto levels similar to in the western hemlock zone,but at a much slower rate. Early stand develop-ment may be initiated following stand-replacingfire (Bekker and Taylor 2010), prolonged patho-gen exposure (Matson and Waring 1984), orexpansion into meadows (Taylor 1995, Zald et al.2012) and can be rapid or protracted. Biomassdevelopment and density may continue toincrease until later stages of development whenpartial stand-replacing disturbance from a rootrot pathogen (laminated root rot, Phellinus weiriiMurr.; Matson and Waring 1984) or stand-replacing wildfire (Agee 1993) may occur. Livebiomass decreases as waves of mortality createlocalized patches of complete stand-replacementover time (Matson and Boone 1984, Hansen andGoheen 2000) or the stand may recover andregain live biomass on a trajectory towards old-growth stages. In the absence of root diseasedevelopment may continue towards old-growth.

Structural variation in western juniper sug-gests that development operates within a verylow range of biomass and density, both of whichare primarily limited by climate. Early develop-mental stages may be initiated following high-severity wildfire (Campbell et al. 2012) orcomplete mechanical removal of live trees (Bateset al. 2005). Expansion into non-forested grass-lands or shrublands, however, has been morecommon over the last century and accounts for a;900% increase in western juniper (Miller andRose 1995). Expansion may be rapid, but growthis slow and establishment is protracted with bigtrees developing prior to canopy closure. Anabrupt shift towards an old-growth stage withclosed canopy (Waichler et al. 2001) may occurfrom infilling associated with changing distur-bance regimes or favorable climate (Soule et al.2004). Old-growth juniper was historically re-stricted to sites offering refuge from fire, but hasencroached on sagebrush steppe over the last few


REILLY AND SPIES

centuries following grazing, decreased fire oc-currence, and relatively moist climatic conditions(Miller and Rose 1999). In the absence ofdisturbance the old-growth stage may persistfor centuries (Romme et al. 2009).

Structural patterns in the grand fir/white firvegetation zone suggest that these developmen-tal pathways are among the most complex in theregion. Early developmental stages may havebeen initiated by stand-replacing fire, logging(Adams and Latta 2007), or expansion intomeadows or shrublands (Halpern et al. 2010).Early developmental stages in the late 20thcentury have been truncated and with rapidcanopy closure following planting for timbermanagement (Seidel 1979, Sensenig et al. 2013),but our data suggest that pathways of protracteddevelopment have also been common. Most latedevelopmental stages are associated with adifferent pathway which includes repeated re-moval of large trees and fire exclusion followedby recruitment and infilling of shade toleranttrees dating to late 19th and early 20th centuries(Youngblood et al. 2004, Merschel et al. 2014) andcurrent basal area and density are much higherthan in the late 19th and early 20th century(Baker 2012, Hagmann et al. 2013, Hagmann etal. 2014). Current pathways in later developmen-tal stages may include mechanical thinning forrestoration and reduction of risk of high-severityfire (Harrod et al. 2009), but natural disturbanceincluding fire (Littell et al. 2009, Wimberly andLiu 2014, Cansler and McKenzie 2014) andinsects (Meigs et al. 2015) have been morecommon in recent years. Some stand-replace-ment fire has occurred in a small proportion ofthis vegetation zone (;0.9%), but partial stand-replacement fire has been more common (1.6%)and both are rare compared to the total extent ofthe vegetation zone in late and old-growthdevelopmental stages (;46%). Old-growth stag-es with extremely high biomass and dead woodmay persist in topographically related firerefugia or other remote areas that were notlogged (Camp 1999), but are uncommon (2.1%).

Implications for conservation and managementOur study represents the first empirical char-

acterization of early developmental stages andcan serve as a reference and for assessing thecontribution of early-seral forests to landscape

diversity and for developing strategies aimed atconserving or creating them. Declines in early-seral forests around the world (Angelstam 1998,Trani et al. 2001) have been linked with corre-sponding losses in biodiversity (Pimm andAskins 1995, Betts et al. 2010) and are a focusof some recent forest restoration proposals in thePacific Northwest (Franklin and Johnson 2012).The large degree of variation in the structure ofearly developmental stages that we found sub-stantiates the need to differentiate among early-seral forests for conservation purposes. Deadbiological legacies have the potential to increasestructural complexity in early developmentalstages and provide important habitat elementsfor a variety of organisms that may not bepresent in all early-seral forests (Hutto 2008,Swanson et al. 2014). In addition, although thescale and frequency of disturbance affects theextent of early successional habitats (Lorimer andWhite 2003), variation in productivity and ratesof seedling establishment and regrowth will alsoaffect the extent of early-seral habitats which maybe ephemeral in productive vegetation zones butpersist for much longer where it is cold or dry.Diverse early-seral forests are still one of therarest habitats in the region, but the amount ofthis habitat varies tremendously with scale anddiffers among vegetation zones and geographicallocations. Most diverse early-seral forests arelocated in a few ‘‘hotspots’’ associated with verylarge wildfires in the Klamath (Biscuit Fire 2002),Eastern Cascades/Cascades (B and B Fire 2003),and North Cascades (Tripod Fire 2007) ecore-gions (Appendix A1).

High density moderate biomass plantationsare a major source of timber in many regions ofthe world (Hartley 2002). Although they are notoften the focus of conservation, their manage-ment has important implications for biodiversitysince they comprise the majority of non-reserveforest lands (Franklin and Lindenmayer 2009)but typically lack habitat elements (e.g., deadwood, large remnant trees) that many speciesmay depend on (Johnson and O’Neil 2001). Amesofilter conservation approach aimed at con-serving specific elements in managed forests(Hunter 2005) could enhance structural diversityand habitat function in managed forests. Othermoderate biomass structural classes with lowerdensity of live trees pertain to conservation issues


REILLY AND SPIES

in drier forests where savannahs and woodlandsaround the world have gone through recentrapid structural change from either mortality(Allen and Breshears 1998, Williams et al. 2010,Vila-Cabrera et al. 2011) or expansion (Weltzinand McPherson 1999, Archer 2010, Gimeno et al.2012) with important consequences for ecosys-tem function (Sala and Maestre 2014) andbiodiversity (Ratajczak et al. 2012). Our resultsindicate that western juniper has experiencedvery little recent change related to mortality, butsome woodland types with big trees and highdensities of snags in the ponderosa pine, Doug-las-fir, and grand fir/white fir vegetation zonesare evidence of recent episodes of mortality thathave affected anywhere from ;8–11 % of thesevegetation zones. Our results suggest that manydry forests in the Pacific Northwest are goingthrough a period of thinning from mortality, butdecreases in overall density and increases thedominance of big trees may be restoring someaspects of historical structure.

The number of structural classes in the highlive biomass group demonstrate the diversity oflater developmental stages across environmentsand vegetation zones in a region that stillsupports millions of hectares of old-growth forest(Moeur et al. 2011). Late successional and old-growth have been the focus of considerableconservation efforts and around the world(Woodgate et al. 1996, Kimmins 2003, Spies andDuncan 2009; Davis et al., in press) due to theirunique contribution to biodiversity and role theyplay in global carbon storage (Luyssaert et al.2008). High densities of big trees and snags andlarge amounts of dead and downed wood areshared attributes of old-growth forests aroundthe world (D’Amato et al. 2008, Burrascano et al.2013), but these attributes may not be present inlate developmental stages of all vegetation zoneswhere environment or disturbance constraindevelopment of one or more components ofstructure (e.g., frequent fire consumes deadwood). Refined definitions based on regionalinventories incorporating the full range of poten-tial structural conditions have the potential toavoid confusion that has occurred in the pastregarding policy and management (Parker et al.2000, Spies 2004). Our results also clearly showthat the relationship between stand age andstructural development does not apply to all

‘‘old’’ forests, especially those in low productivityor frequently disturbed environments (O’Hara etal. 1996). The general model of stand develop-ment from the western hemlock zone thatemphasizes development of high live and deadbiomass and canopy layering is not necessarily agood basis for assessment and conservation of‘‘older’’ forest in other vegetation zones. Rates ofdevelopment differ along regional gradients inproductivity and complex age structures canresult from both partial stand-replacing distur-bance and protracted development dry or coldenvironments. Finally, despite the numerousecological roles of forests in later developmentalstages, perhaps one of the least recognized is thatthey are the only source of the unique habitatelements capable of creating highly diverse early-seral habitats.

ConclusionThis study provides the first empirically based

regional-scale characterization of forest structureand how it varies in relation to age, vegetationzone, and disturbance history. Although most ofvariation in forest structure was accounted for bylive tree biomass, other components related totree size and density, dead wood, and understoryvegetation cumulatively accounted for as muchvariation. Our results indicate that structuraldevelopment across regional gradients in envi-ronment, species composition, and disturbancehistory include more complex pathways than themonotonic accumulation of live biomass oftendepicted in simple conceptual models of standdevelopment following stand-replacing distur-bance in relatively productive environments.Indicative of the variation in importance andtiming of the various ecological processes oper-ating during development, some components ofstructure may act independently or emerge atmultiple points during development. As a result,models of structural development based only onlive biomass are inadequate to account fordynamics at a regional scale.

The regional scale classification of structuraltypes reveals the diversity of ways that interac-tions among individual structural componentsmay be expressed at the stand-scale. We identi-fied twenty-five structural types with uniquecombinations of live tree biomass, density, andsize, snag and dead wood abundance, and


REILLY AND SPIES

understory cover. High levels of structuraldiversity occurred at all levels of live biomass,but diverse low biomass structural types were farless common than diverse types with high livebiomass. Although age typically increased withbiomass across structural classes in wet forests,the relationship between age and structural classwas much weaker in dry forests where a historyof mixed-severity fire regimes and pathogen andinsect outbreaks have created a complex mosaicof forest structure and wide range of tree ages ona site.

Most structural types corresponded well withstages described in current models of standdevelopment, but some types associated withprotracted early development, transitions be-tween woodland/savannah and closed canopiedforest conditions, and partial stand-replacingdisturbance substantiate the need for a broaderframework for conceptualizing structural devel-opment and tracking regional forest dynamics.Structural development in woodlands and sa-vannahs includes many of the same processes(e.g., self-thinning, infilling or understory re-initiation) that occur in closed canopy forests, butthey may occur repeatedly or at different times.Development may be slow, but transitions canhappen rapidly as infilling trees approach per-colation thresholds for canopy closure (Abades etal. 2014) or partial stand-replacing disturbancesreduce density of smaller trees. Thus, woodlandsand savannah dynamics can be characterized bya persistent structural backbone of big treeswhere the density of smaller trees fluctuatesdepending on climatic conditions and the fre-quency and severity of disturbance. We proposeseveral general models using a multiple pathwayframework based on interpretation of structuralpatterns, recent fire history, and general knowl-edge of disturbance and management historyacross the region. General models can bedistinguished based on the range of variationalong various components of structure (e.g.,biomass and density), the rate and timing oftransitions among stages, and complexity ofpathways and relative importance of differentdisturbance types. These models can serve as aconceptual framework for developing morecomprehensive models of structural develop-ment that apply to a wider variety of vegetationzones with different environments and distur-

bance regimes.

Future research and applicationOur study provides a complimentary approach

to the traditional chronosequence approach ofsampling of a small number of subjectivelyselected stands. Despite incorporating and char-acterizing the wide range of structural conditionsacross a diverse region, our inferences regardingecological processes are based primarily onobservations of structural patterns and knowl-edge of management and disturbance historywith support from a rich literature on dynamicsof forests, woodlands, and savannahs around theworld. There is a need for future studies toexplicitly test many of our findings, particularlythose regarding the role of mortality in variousdevelopmental stages. Intensive plot-based stud-ies have the potential to elucidate the roles ofthese processes by explicitly quantifying themand incorporating finer scale spatial pattern toprovide a more mechanistic understanding oftheir consequence on structural development.Proposed conceptual models of structural devel-opment can be tested in other vegetation zones toassess their generality and limitations. Finally,this study provides a reference and frameworkwith which to evaluate historical and potentialfuture trajectories of landscape diversity relatedto changes in climate, disturbance, and manage-ment.

ACKNOWLEDGMENTS

Funding for this work was provided by the USDAForest Service Region 6 Inventory and Analysisprogram. We thank Heather Roberts, Andy Gray, MattGregory, and Janet Ohmann for their assistance inmanaging and summarizing inventory data. We alsothank Bruce McCune for sharing his insights on theanalysis approach as well as Rob Pabst, Harold Zald,and Andrew Merschel for discussions regardinginterpretation of the classification. James Johnstonand Dave Hibbs provided feedback on early versionsof the manuscript and Ray Davis, Julia Burton, JanetOhmann, John Bailey, Matt Betts, Patricia Muir, DavidL. Peterson, and two anonymous reviewers providedextensive critiques which greatly improved the finalversion of manuscript. We thank Dave Bell forassistance making figures and Kathryn Ronnenbergfor assistance with graphic design.


REILLY AND SPIES

LITERATURE CITED

Abades, S. R., A. Gaxiola, and P. A. Marquet. 2014.Fire, percolation thresholds and the savannahforest transition: a neutral model approach. Journalof Ecology 102:1386–1393.

Acacio, V., M. Holmgren, P. A. Jansen, and O.Schrotter. 2007. Multiple recruitment limitationcauses arrested succession in Mediterranean corkoak systems. Ecosystems 10:1220–1230.

Adams, D. M., and G. S. Latta. 2007. Future prospectsfor private timber harvest in Eastern Oregon.Western Journal of Applied Forestry 22:197–203.

Agee, J. K. 1993. Fire ecology of Pacific Northwestforests. Island Press, Washington, D.C., USA.

Agee, J. K., and L. Smith. 1984. Subalpine treereestablishment after fire in the Olympic Moun-tains, Washington. Ecology 65:810–819.

Allen, C. D., and D. D. Breshears. 1998. Drought-induced shift of forest-woodland ecotone: Rapidlandscape response to climate variation. Proceed-ings of the National Academy of Sciences USA95:14839–14842.

Angelstam, P. K. 1998. Maintaining and restoringbiodiversity in European boreal forests by devel-oping natural disturbance regimes. Journal ofVegetation Science 9:593–602.

Archer, S. 2010. Rangeland conservation and shrubencroachment: new perspectives on an old prob-lem. Pages 53–97 in J. du Toit, R. Kock, and J.Deutsch, editors. Wild rangelands: conservingwildlife while maintaining livestock in semi-aridecosystems. Wiley-Blackwell, Oxford, UK.

Ares, A., C. Bright, and K. Puettmann. 2012. Mesoscalevariation in snag and hardwood densities and sizesin old-growth forests in Western Oregon. WesternJournal of Applied Forestry 27:12–17.

Aubry, K. B., C. B. Halpern, and C. E. Peterson. 2009.Variable-retention harvests in the Pacific North-west: a review of short-term findings from theDEMO study. Forest Ecology and Management258:398–408.

Australian and New Zealand Environment Conserva-tion Council. 2000. National Framework for theManagement and Monitoring of Australia’s NativeVegetation. Environment Australia, Canberra, Aus-tralia.

Bailey, J. D., and J. C. Tappeiner. 1998. Effects ofthinning on structural development in 40- to 100-year-old Douglas-fir stands in western Oregon.Forest Ecology and Management 108:99–113.

Baker, W. L. 2012. Implications of spatially extensivehistorical data from surveys for restoring dryforests of Oregon’s eastern Cascades. Ecosphere3:23. http://dx.doi.org/10.1890/ES11-00320.1

Barbour, M. G., and J. Major. 1988. Terrestrialvegetation of California. California Native Plant

Society, Sacramento, California, USA.Bates, J. D., R. F. Miller, and T. Svejcar. 2005. Long-term

successional trends following western junipercutting. Rangeland Ecology and Management58:533–541.

Bekker, M. F., and A. H. Taylor. 2010. Fire, disturbance,forest structure, and stand dynamics in montaneforests of the southern Cascades, Thousand LakesWilderness, California, USA. Ecoscience 17:59–72.

Berkowitz, A. R., C. D. Canham, and V. R. Kelly. 1995.Competition vs. facilitation of tree seedling growthand survival in early successional communities.Ecology 76:1156–1168.

Betts, M. G., J. C. Hagar, J. W. Rivers, J. D. Alexander,K. McGarigal, and B. C. McComb. 2010. Thresholdsin forest bird occurrence as a function of theamount of early-seral broadleaf forest at thelandscape scale. Ecological Applications 20:2116–2130.

Bormann, F. H., and G. E. Likens. 1979. Pattern andprocess in a forested ecosystem. Springer, NewYork, New York, USA.

Brown, P. M., and R. Wu. 2005. Climate anddisturbance forcing of episodic tree recruitment ina southwestern ponderosa pine landscape. Ecology86:3030–3038.

Brown, W. C. 1968. Early Okanogan history. YeGalleon Press, Seattle, Washington, USA. S. Bur-rascano, W. S. Keeton, F. M. Sabatini, and C. Blasi.2013. Commonality and variability in the structuralattributes of moist temperate old-growth forests: aglobal review. Forest Ecology and Management291:459–479.

Burrascano, S., W. S. Keeton, F. M. Sabatini, and C.Blasi. 2013. Commonality and variability in thestructural attributes of moist temperate old-growthforests: a global review. Forest Ecology andManagement 291:459–479.

Calder, W. J., and S. B. St. Clair. 2012. Facilitationdrives mortality patterns along succession gradi-ents of aspen-conifer forests. Ecosphere 3:57. http://dx.doi.org/10.1890/ES12-00119.1

Cale, W. G., G. M. Henebry, and J. A. Yeakley. 1989.Inferring process from pattern in natural commu-nities. BioScience 39:600–605.

Callaway, R. M. 1998. Competition and facilitation onelevation gradients in subalpine forests of thenorthern Rocky Mountains, USA. Oikos 82:561–573.

Camp, A. E. 1999. Age structure and species compo-sition resulting from altered disturbance regimeson the eastern slope of the Cascades Range,Washington. Journal of Sustainable Forestry 9:39–67.

Campbell, J. L., R. E. Kennedy, W. B. Cohen, and R. F.Miller. 2012. Assessing the carbon consequences ofwestern juniper (Juniperus occidentalis) encroach-


REILLY AND SPIES

ment across Oregon, USA. Rangeland Ecology andManagement 65:223–231.

Cansler, A., and D. McKenzie. 2014. Climate, fire size,and biophysical setting control fire severity andspatial pattern in the Northern Cascade Range,USA. Ecological Applications 24:1037–1056.

Chambers, J. M., W. S. Cleveland, B. Kleiner, and P. A.Tukey. 1983. Graphical methods for data analysis.Wadsworth, Monterey, California, USA.

Chan, S. S., D. J. Larson, K. G. Maas-Hebner, W. H.Emmingham, S. R. Johnston, and D. A. Mikowski.2006. Overstory and understory development inthinned and underplanted Oregon Coast RangeDouglas-fir stands. Canadian Journal of ForestResearch 36:2696–2711.

Crookston, N. L., and A. R. Stage. 1999. Percentcanopy cover and stand structural statistics fromthe Forest Vegetation Simulator. RMRS GTR-24.USDA Forest Service, Rocky Mountain ResearchStation. Ogden, Utah, USA.

Curtis, R. O., D. S. DeBell, C. A. Harrington, D. P.Lavender, J. B. St. Claire J. C. Tappeiner, and J. D.Walstad. 1998. Silviculture for multiple objectivesin the Douglas-fir region. PNW GTR-435. USDAForest Service, Pacific Northwest Research Station,Portland, Oregon, USA.

D’Amato, A. W., D. A. Orwig, and D. R. Foster. 2008.The influence of successional processes and distur-bance on the structure of Tsuga canadensis forests.Ecological Applications 18:1182–1199.

Davis, R. J., J. L. Ohmann, R. E. Kennedy, W. B. Cohen,M. J. Gregory, Z. Yang, H. M. Roberts, A. N. Gray,and T. A. Spies. In press. Northwest Forest Plan–thefirst 20 years (1994–2013): status and trends of late-successional and old-growth forests. PNW GTR-911. Portland, Oregon: U.S. Department of Agri-culture, Forest Service, Pacific Northwest ResearchStation.

Donato, D. C., J. L. Campbell, and J. F. Franklin. 2012.Multiple successional pathways and precocity inforest development: Can some forests be borncomplex? Journal of Vegetation Science 23:576–584.

Franklin, J. F., and C. T. Dyrness. 1973. Naturalvegetation of Oregon and Washington. OregonState University Press, Corvallis, Oregon, USA.

Franklin, J. F., and N. K. Johnson. 2012. A restorationframework for federal forest in the Pacific North-west. Journal of Forestry 110:429–439.

Franklin, J. F., and D. B. Lindenmayer. 2009. Impor-tance of matrix habitats in maintaining biologicaldiversity. Proceedings of the National Academy ofSciences USA 106:349–350.

Franklin, J. F., and R. Van Pelt. 2004. Spatial aspects ofstructural complexity in old-growth forests. Journalof Forestry 102:22–28.

Franklin, J. F., et al. 2002. Disturbances and structuraldevelopment of natural forest ecosystems with

silvicultural implications, using Douglas-fir forestsas an example. Forest Ecology and Management155:399–423.

Frelich, L. E. 2002. Forest dynamics and disturbanceregimes: studies from the temperate-evergreendeciduous forests. Cambridge University Press,New York, New York, USA.

George, L. O., and F. A. Bazazz. 1999. The fernunderstory as an ecological filter: emergence andestablishment of canopy-tree seedlings. Ecology80:833–845.

Gholz, H. L. 1982. Environmental limits on above-ground net primary production, leaf area, andbiomass in the vegetation zone of the PacificNorthwest. Ecology 63:469–481.

Gimeno, T. E., A. Escudero, A. Delgado, and F.Valladares. 2012. Previous land use alters the effectof climate change on and facilitation on expandingwoodlands of Spanish juniper. Ecosystems 15:564–579.

Gray, A. N., and T. A. Spies. 1996. Gap size, within-gapposition, and canopy structure effects on coniferseedling establishment in forest gaps. Ecology84:635–645.

Hagmann, R. K., J. F. Franklin, and K. N. Johnson.2013. Historical structure and composition ofponderosa pine and mixed-conifer forests insouth-central Oregon. Forest Ecology and Manage-ment 304:492–504.

Hagmann, R. K., J. F. Franklin, and K. N. Johnson.2014. Historical conditions in mixed-conifer forestson the eastern slopes of the northern OregonCascade Range, USA. Forest Ecology and Manage-ment 330:158–170.

Halpern, C. B. 1988. Early successional pathways andthe resistance and resilience of forestcommunities.Ecology 69:1703–1715.

Halpern, C. B., J. A. Antos, J. M. Rice, and R. Haugo.2010. Tree invasion of a montane meadow com-plex: temporal trends, spatial patterns, and bioticinteractions. Journal of Vegetation Science 21:717–732.

Hansen, A. J., T. A. Spies, F. J. Swanson, and J. L.Ohmann. 1991. Conserving biodiversity in man-aged forests. BioScience 41:382–392.

Hansen, E. M., and E. M. Goheen. 2000. Phellinus weiriiand other native root pathogens as determinants offorest structure and process in western NorthAmerica. Annual Review of Phytopathology38:515–539.

Harmon, M. E., et al. 1986. The ecology of coarsewoody debris in temperate ecosystems. Advancesin Ecological Research 15:133–302.

Harrod, R. J., D. W. Peterson, N. A. Povak, and E. K.Dodson. 2009. Thinning and prescribed fire effectson overstory tree and snag structure in dryconiferous forests of the interior Pacific Northwest.


REILLY AND SPIES

Forest Ecology and Management 258:712–721.Hartley, M. J. 2002. Rationale and methods for

conserving biodiversity in plantation forests. ForestEcology and Management 155:81–95.

Healey, S. P., W. B. Cohen, T. A. Spies, M. Moeur, D.Pflugmacher, M. G. Whitely, and M. Lefskey. 2008.The relative impact of harvest and fire uponlandscape-level dynamics of older forests: lessonsfrom the Northwest Forest Plan. Ecosystems11:1106–1119.

Hemstrom, M. A., A. A. Ager, M. Vavra, B. C. Wales,and M. J. Wisdom. 2004. A state and transitionapproach for integrating landscape models. Pages17–32 in J. L. Hayes, A. A. Ager, and R. J. Barbour,editors. Methods for integrated modeling of land-scape change: interior Northwest landscape analy-sis system. PNW GTR-610. USDA Forest Service,Pacific Northwest Research Station, Portland, Ore-gon, USA.

Hessburg, P. F., and J. K. Agee. 2003. An environmentalnarrative of Inland Northwest United States for-ests, 1800-2000. Forest Ecology and Management178:23–59.

Hudiburg, T., B. Law, D. P. Turner, J. Campbell, D.Donato, and M. Duane. 2009. Carbon dynamics ofOregon and Northern California forests andpotential land-based storage. Ecological Applica-tions 19:163–180.

Hunter, M. L. 2005. A mesofilter conservation strategyto complement fine and coarse filters. ConservationBiology 19:1025–1029.

Hutto, R. L. 2008. The ecological importance of severewildfires: some like it hot. Ecological Applications18:1827–1834.

Johnson, D. H., and T. A. O’Neil. 2001. Wildlife-habitatrelationships in Oregon and Washington. OregonState University Press, Corvallis, Oregon, USA.

Johnson, K. N., and F. J. Swanson. 2009. Historicalcontext of old-growth forests in the Pacific North-west-policy, practices, and competing worldviews.Pages 12–28 in T. A. Spies and S. L. Duncan,editors. Old growth in a new world: a PacificNorthwest icon reexamined. Island Press, Wash-ington, D.C., USA.

Kimmins, J. P. 2003. Old-growth forest: An ancient andstable sylvan equilibrium, or a relatively transitoryecosystem condition that offers people a visual andemotional feast? Answer: it depends. ForestryChronicle 79:429–440.

Langston, N. 1995. Forest Dreams, forest nightmares:the paradox of old growth in the inland West.University of Washington Press, Seattle, Washing-ton, USA.

Littell, J. S., D. McKenzie, D. L. Peterson, and A. L.Westerling. 2009. Climate and wildfire burned inthe western U.S. ecoprovinces, 1916-2003. Ecolog-ical Applications 19:1003–1021.

Lorimer, C. G., and A. S. White. 2003. Scale andfrequency of natural disturbances in the northeast-ern US: implications for early successional foresthabitats and regional age distributions. ForestEcology and Management 185:41–64.

Lutz, J. A., and C. B. Halpern. 2006. Tree mortalityduring early forest development: a long-term studyof rates, causes, and consequences. 76:257–275.

Lutz, J. A., A. J. Larson, T. J. Furniss, D. C. Donato, J. A.Freund, M. E. Swanson, K. J. Bible, J. Chen, and J. F.Franklin. 2014. Spatially nonrandom tree mortalityand ingrowth maintain equilibrium pattern in anold-growth Pseudotsuga-Tsuga forest. Ecology95:2047–2054.

Luyssaert, S., E. Detlef Schultz, A. Borner, A. Knohl, D.Hessenmoller, B. E. Law, P. Ciais, and J. Grace.2008. Old-growth forests as global carbon sinks.Nature 455:213–215.

MacArthur, R. H., and J. W. MacArthur. 1961. On birdspecies diversity. Ecology 42:594–598.

Malhi, Y., et al. 2006. The regional variation ofaboveground live biomass in old-growth Amazo-nian forests. Global Change Biology 12:1107–1138.

Matson, P. A., and R. D. Boone. 1984. Naturaldisturbance and nitrogen mineralization: wave-form dieback of mountain hemlock in the OregonCascades. Ecology 65:1511–1516.

Matson, P. A., and R. W. Waring. 1984. Effects ofnutrient and light limitation on mountain hemlock:susceptibility to laminated root rot. Ecology65:1517–1524.

McComb, B. C. 2008. Wildlife habitat management:concepts and applications in forestry. CRC Press/Taylor and Francis Group, Boca Raton, Florida,USA.

McComb, W. C., T. A. Spies, and W. H. Emmingham.1993. Douglas-fir forests: managing for timber andmature-forest habitat. Journal of Forestry 91:31–42.

McCune, B., and T. F. H. Allen. 1985. Will similarforests develop on similar sites? Canadian Journalof Botany 63:367–376.

Meigs, G. W., R. E. Kennedy, A. N. Gray, and M.Gregory. 2015. Spatiotemporal dynamics of recentmountain pine beetle and western spruce bud-worm outbreaks across the Pacific NorthwestRegion, USA. Forest Ecology and Management339:71–86.

Merschel, A., T. A. Spies, and E. Heyerdahl. 2014.Mixed-conifer forests of Central Oregon: effects oflogging and fire exclusion vary with environment.Ecological Applications. 24:1670–1688.

Miller, R. F., and J. R. Rose. 1995. Historic expansion ofJuniperus occidentalis (western juniper) in southeastOregon. Great Basin Naturalist 55:37–45.

Miller, R. F., and J. R. Rose. 1999. Fire history andwestern juniper encroachment in sage-brushsteppe. Journal of Range Management 52:550–559.


REILLY AND SPIES

Montreal Process Working Group. 2009. Criteria andindicators for the conservation of and sustainabilitymanagement of temperate and boreal forests.http://www.montreal process.org/Resources/Publi-cations/index.shtml

Moeur, M., J. L. Ohmann, R. E. Kennedy, W. B. Cohen,M. J. Gregory, Z. Yang, H. M. Roberts, T. A. Spies,and M. Fiorella. 2011. Northwest Forest Plan—thefirst fifteen years (1994-2008): status and trends oflate-successional and old-growth forests. PNWGTR-853. USDA Forest Service, Pacific NorthwestResearch Station, Portland, Oregon, USA.

Muir, P. S. 1993. Disturbance effects on structure andtree species composition of Pinus contorta forests inwestern Montana. Canadian Journal of ForestResearch 23:1617–1625.

O’Hara, K. L., P. A. Latham, P. Hessburg, and B. G.Smith. 1996. A structural classification for inlandnorthwest forest vegetation. Western Journal ofApplied Forestry 11:97–102.

Ohmann, J. L., M. J. Gregory, and T. A. Spies. 2007.Influence of environment, disturbance, and own-ership on forest vegetation of coastal Oregon.Ecological Applications 17:18–33.

Ohmann, J. L., and T. A. Spies. 1998. Regional gradientanalysis and spatial pattern of woody plantcommunities of Oregon forests. Ecological Mono-graphs 68:151–182.

Ohmann, J. L., and K. L. Waddell. 2002. Regionalpatterns of dead wood in forested habitats ofOregon and Washington. Pages 535–560 in W. F.Laudenslayer, P. J. Shea, B. E. Valentine, C. P.Weatherspoon, and T. E. Lisle, technical coordina-tors. Proceedings of the symposium on the ecologyand management of dead wood in western forests,Reno, Nevada November 2–4, 1999. PSW GTR-181.USDA Forest Service, Pacific Southwest ResearchStation, Albany, California, USA.

Oliver, C. D., and B. C. Larson. 1990. Forest standdynamics. McGraw-Hill, New York, New York,USA.

Parker, A. J., K. C. Parker, and H. Wiggins-Brown.2000. Disturbance and scale effects on southernold-growth forests (USA): the sand pine example.Natural Areas Journal 20:273–279.

Perry, D. A., P. F. Hessburg, C. N. Skinner, T. A. Spies,S. L. Stephens, A. H. Taylor, J. F. Franklin, B.McComb, and G. Riegel. 2011. The ecology ofmixed severity fires regimes in Washington, Ore-gon, and Northern California. Forest Ecology andManagement 262:703–717.

Pimm, S. L., and R. A. Askins. 1995. Forest lossespredict bird extinctions in eastern North America.Proceedings of the National Academy of SciencesUSA 92:9343–9347.

Platt, W. J., G. W. Evans, and S. L. Rathbun. 1988. Thepopulation dynamics of a long-lived conifer (Pinus

palustris). American Naturalist 131:491–525.Pommerening, A. 2002. Approaches to quantifying

forest structures. Forestry 75:305–324.Putz, F. E., and C. D. Canham. 1992. Mechanisms of

arrested succession in shrublands: root and shootcompetition between shrubs and tree seedlings.Forest Ecology and Management 49:267–275.

Ratajczak, Z., J. B. Nippert, and S. L. Collins. 2012.Woody encroachment decreases diversity acrossNorth American grasslands and savannahs. Ecol-ogy 93:697–703.

R Development Core Team. 2011. R: a language andenvironment for statistical computing. R Founda-tion for Statistical Computing, Vienna, Austria.

Rice, J. M., C. B. Halpern, J. A. Antos, and J. J. Jones.2012. Spatio-temporal patterns of tree establish-ment are indicative of biotic interactions duringearly invasion of a montane meadow. PlantEcology 213:555–568.

Riginos, C. 2009. Grass competition suppresses savan-nah tree growth across multiple demographicstages. Ecology 90:335–340.

Romme, W. H., et al. 2009. Historical and moderndisturbance regimes, stand structures and land-scape dynamics in Pinon-Juniper vegetation of thewestern United States. Rangeland Ecology andManagement 62:203–222.

Rooney, T. M., and D. Waller. 2003. Direct and indirecteffects of white-tailed deer browsing in forestecosystems. Forest Ecology and Management181:165–176.

Sala, O. E., and F. T. Maestre. 2014. Grass-woodlandtransitions: determinants and consequences forecosystem functioning and provision of services.Journal of Ecology 102:1357–1362.

Seidel, K. W. 1979. Regeneration in mixed coniferclearcuts in the Cascade Range and Blue Moun-tains of eastern Oregon. PNW GTR-248. USDAForest Service, Pacific Northwest Forest and RangeExperimental Station, Portland, Oregon, USA.

Sensenig, T., J. D. Bailey, and J. C. Tappeiner. 2013.Stand development, fire and growth of old-growthand young forests in southwestern Oregon, USA.Forest Ecology and Management 291:96–109.

Simpson, M. 2013. Developer of the forest vegetationzone map. Ecologist, Central Oregon Area Ecologyand Forest Health Program. USDA Forest Service,Pacific Northwest Region, Bend, Oregon, USA.

Sinton, D. S., J. A. Jones, J. L. Ohmann, and F. J.Swanson. 2000. Windthrow disturbance, forestcomposition, and structure in the Bull Run Basin,Oregon. Ecology 81:2539–2556.

Soule, P. T., P. A. Knapp, and H. D. Grissino-Mayer.2004. Human agency, environmental drivers, andwestern juniper establishment during the lateHolocene. Ecological Applications 14:96–112.

Spies, T. A. 1998. Forest structure: a key to the


REILLY AND SPIES

ecosystem. Northwest Science 72:34–39.Spies, T. A. 2004. Ecological concepts and diversity of

old-growth forests. Journal of Forestry 102:14–20.Spies, T. A., and S. L. Duncan. 2009. Old growth in a

new world: a Pacific Northwest icon reexamined.Island Press, Washington, D.C., USA.

Spies, T. A., and J. F. Franklin. 1991. The structure ofnatural young, mature, and old-growth Douglas-firforests in Oregon and Washington. Pages 91–109 inL. F. Ruggerio, K. B. Aubry, A. B. Carey, and M. H.Huff, editors. Wildlife and vegetation in unman-aged Douglas-fir forests. PNW GTR-285. USDAForest Service Pacific Northwest Research Station,Portland, Oregon, USA.

Spies, T. A., J. F. Franklin, and T. B. Thomas. 1988.Coarse woody debris in Douglas-fir forests ofwestern Oregon and Washington. Ecology69:1689–1702.

Spies, T. A., B. C. McComb, R. S. H. Kennedy, M. T.McGrath, K. Olsen, and R. J. Pabst. 2007. Potentialeffects of forest policies on terrestrial biodiversity ina multi-ownership province. Ecological Applica-tions 17:48–65.

Staver, A. C., S. Archibald, and S. A. Levin. 2011. Theglobal extent and determinants of savanna andforest as alternative biome states. Science 334:230–232.

Swanson, M. E., J. F. Franklin, R. L. Beschta, C. M.Crisafulli, D. A. DellaSala, R. L. Hutto, D. B.Lindenmayer, and F. J. Swanson. 2010. Theforgotten stage of forest succession: early succes-sional ecosystems on forest sites. Frontiers inEcology and the Environment 9:117–125.

Swanson, M. E., N. M. Studevant, J. L. Campbell, andD. C. Donato. 2014. Biological associates of early-seral pre-forest in the Pacific Northwest. ForestEcology and Management 324:160–171.

Taylor, A. H. 1995. Forest expansion and climatechange in the mountain hemlock zone (Tsugamertensiana) zone, Lassen Volcanic National Park,California, USA. Arctic and Alpine Research27:207–216.

Taylor, A. H. 2010. Fire disturbance and foreststructure in an old-growth Pinus ponderosa forest,southern Cascades, USA. Journal of VegetationScience 21:561–572.

Taylor, A. H. and C. N. Skinner. 2003. Spatial patternsand controls on historical fire regimes and foreststructure in the Klamath Mountains. EcologicalApplications 13:704–719.

Tepley, A. J., F. J. Swanson, and T. A. Spies. 2013. Fore-mediated pathways of stand development inDouglas-fir/western hemlock forest of the PacificNorthwest, USA. Ecology 94:1729–1743.

Tepley, A. J., F. J. Swanson, and T. A. Spies. 2014. Post-fire tree establishment and early cohort develop-ment in conifer forests of the Western Cascades of

Oregon, USA. Ecosphere 5:art80.Trani, M. K., R. T. Brooks, T. L. Schmidt, V. A. Rudis,

and C. M. Gabbard. 2001. Patterns and trends ofearly-successional forests in the eastern UnitedStates. Wildlife Society Bulletin 29:413–424.

Urquiza-Haas, T., P. M. Dolman, and C. A. Peres. 2007.Regional scale variation in forest structure andbiomass in the Yucatan Peninsula, Mexico: effectsof forest disturbance. Forest Ecology and Manage-ment 247:80–90.

USDA Forest Service, DOI [Department of the Interi-or], and BLM [Bureau of Land Management]. 1994.Standards and guidelines for management ofhabitat for late-successional and old-growth forestrelated species within the range of the NorthernSpotted Owl. USDA Forest Service, Portland,Oregon, USA and BLM, Moscow, Idaho, USA.

U.S. Fish and Wildlife Service. 2003. Revised recoveryplan for the red-cocked woodpecker (Picoidesborealis). U.S. Fish and Wildlife Service, Atlanta,Georgia, USA.

U.S. Fish and Wildlife Service. 2011. Revised recoveryplan for the Northern Spotted Owl (Strix occidenta-lis caurina). U.S. Fish and Wildlife Service, Portland,Oregon, USA.

Veblen, T. T., K. S. Hadley, M. S. Reid, and A. J.Rebertus. 1991. The response of subalpine forests tospruce beetle outbreak in Colorado. Ecology72:213–231.

Verschuyl, J. P., A. J. Hansen, D. B. McWethy, R.Sallabanks, and R. L. Hutto. 2008. Is the effect offorest structure on bird diversity modified by forestproductivity? Ecological Applications 1155-1170.

Vila-Cabrera, A., J. Martinez-Vilalta, J. Vayreda, and J.Retana. 2011. Structural and climatic determinantsof demographic rates of Scots pine forests acrossthe Iberian Peninsula. Ecological Applications21:1162–1172.

Waichler, W. S., R. F. Miller, and P. S. Doescher. 2001.Community characteristics of old-growth westernjuniper woodlands. Journal of Range Management54:518–527.

Ward, J. H. 1963. Hierarchical grouping to optimize anobjective function. Journal of the American Statis-tical Association 58:236–244.

Waring, R. H., and S. W. Running. 2007. Forestecosystems analysis at multiple scales. AcademicPress, Burlington, Massachusetts, USA.

Weisberg, P. J., and F. J. Swanson. 2003. Regionalsynchroneity in fire regimes of western Oregon andWashington, USA. Forest Ecology and Manage-ment 172:17–28.

Weltzin, J. F., and G. R. McPherson. 1999. Facilitationof conspecific seedling recruitment and shifts intemperate savannah ecotones. Ecological Mono-graphs 69:513–534.

Williams, A. P., C. D. Allen, C. I. Millar, T. W.


REILLY AND SPIES

Swetnam, J. Michaelsen, C. J. Still, and S. W.Leavitt. 2010. Forest response to increasing aridityand warmth in the southwestern United States.Proceedings of the National Academy of Sciences107:21289–21294.

Wimberly, M. C., and Z. Liu. 2014. Interactions ofclimate, fire, and management in future forests ofthe Pacific Northwest. Forest Ecology and Man-agement 327:270–279.

Wimberly, M. C., and T. A. Spies. 2001. Influences ofenvironment and disturbance on forest patterns incoastal Oregon watersheds. Ecology 82:1443–1459.

Winter, L. E., L. B. Brubaker, J. F. Franklin, and E. A.Miller. 2002. Initiation of an old-growth Douglas-firstand in the Pacific Northwest: a reconstructionfrom tree-ring records. Canadian Journal of ForestResearch 32:1039–1056.

Winthers, E., et al. 2005. Terrestrial Ecological Unit

Inventory technical guide. USDA Forest Service,

Washington Office, Ecosystem Management Coor-

dination Staff. Washington, D.C., USA.

Woodgate, P. W., B. D. Peel, J. E. Coram, S. J. Farrell,

K. T. Ritman, and A. Lewis. 1996. Old-growth

forest studies in Victoria, Australia concepts and

principles. Forest Ecology and Management 85:79–

94.

Youngblood, A., T. Max, and K. Coe. 2004. Stand

structure in eastside old-growth ponderosa pine

forests of Oregon and northern California. Forest

Ecology and Management 199:191–217.

Zald, H. S. J., T. A. Spies, M. Huso, and D. Gatziolis.

2012. Climatic, landform, microtopographic, and

overstory canopy controls of tree invasion in a

subalpine meadow landscape, Oregon Cascades,

USA. Landscape Ecology 27:1197–1212.

SUPPLEMENTAL MATERIAL

ECOLOGICAL ARCHIVES

Appendices A–D are available online: http://dx.doi.org/10.1890/ES14-00469.1.sm


REILLY AND SPIES

http://dx.doi.org/10.1890/ES14-00469.1.sm

Regional variation in stand structure and development in ...

Documents